Unit:
Κατεύθυνση Πυρηνική Φυσική και Φυσική Στοιχειωδών ΣωματιδίωνLibrary of the School of Science
Supervisors info:
Νίκη Σαουλίδου, Αναπλ. Καθηγητής Τμήμα Φυσικής, ΕΚΠΑ (Κύρια Επιβλέπουσα),
Θεόδωρος Μερτζιμέκης, Αναπλ. Καθηγητής Τμήμα Φυσικής, ΕΚΠΑ,
Βασίλης Σπανός, ΕΚΠΑ, Αναπλ. Καθηγητής Τμήμα Φυσικής, ΕΚΠΑ
Original Title:
Ταυτοποίηση αδρονικών καταιγισμών από κουάρκς ελαφριάς γεύσης σε αναζητήσεις νέας φυσικής με τέσσερις αδρονικούς πίδακες στην τελική κατάσταση χρησιμοποιώντας δεδομένα Monte Carlo από το πείραμα CMS στον LHC του CERN
Translated title:
Light-quark jet tagging for new physics searches in events with four jets in the final state using Monte Carlo data from the CMS Experiment at the CERN LHC
Summary:
This thesis examines the identification of light-quark jet tagging in events with four hadronic jets in the final state, which are grouped in pairs, in order to search for new physics. For this purpose, we exploit and study an innovative classification utilizing machine learning techniques (Particle Transformer), which incorporates the interactions between pairs of particles in the final training mechanism. This approach allows superior performance over previous techniques. The data come from Monte Carlo simulations of CERN’s Large Hadron Collider (LHC) CMS experiment, in proton-proton collisions at 13.6 TeV. The study focuses on new physics signals involving two up quarks and two gluons in the final state, appearing as resonances at different invariant masses of two and four hadronic jets. With these we study and evaluate the effect of light-quark tagging on the significance of the signals, using both the truth-level information and the reconstructed information from Particle Transformer. The results so far indicate that light jet tagging, using the Particle Transformer, does not increase the discovery potential. In contrast, the use of truth-level information leads to an increase in significance for all signals, especially at lower invariant masses. This suggests that there is room for significant improvements in the performance of jet tagging in the future, with consequent improvement in the potential of discovering new resonances.
Main subject category:
Science
Keywords:
Experimental High Energy Physics, Particle Physics, CMS experiment
File:
File access is restricted only to the intranet of UoA.
Master_thesis_Aikaterini_Karmi.pdf
5 MB
File access is restricted only to the intranet of UoA.
